Evidence for functional protein interactions required for poliovirus RNA replication

Abstract

Poliovirus protein 2C contains a predicted N-terminal amphipathic helix that mediates association of the protein with the membranes of the viral RNA replication complex. A chimeric virus that contains sequences encoding the 18-residue core from the orthologous amphipathic helix from human rhinovirus type 14 (HRV14) was constructed. The chimeric virus exhibited defects in viral RNA replication and produced minute plaques on HeLa cell monolayers. Large plaque variants that contained mutations within the 2C-encoding region were generated upon subsequent passage. However, the majority of viruses that emerged with improved growth properties contained no changes in the region encoding 2C. Sequence analysis and reconstruction of genomes with individual mutations revealed changes in 3A or 2B sequences that compensated for the HRV14 amphipathic helix in the polio 2C-containing proteins, implying functional interactions among these proteins during the replication process. Direct binding between these viral proteins was confirmed by mammalian cell two-hybrid analysis.

FIG. 1.

Schematic diagrams of the chimeric PV genome encoding the N-terminal amphipathic helix in protein 2C from HRV14 and pseudorevertants. (A) The position of the sequence encoding the amphipathic helix is indicated by a black box, and the encoded amino acid sequence is shown below. Amino acids different from those present in the PV sequence are shown in bold. 5′NTR, 5′ nontranslated region. (B) The positions of mutations found in different isolates of viruses with improved growth properties are indicated by arrows. Mutant virus isolates R1 through R5 indicated below the bar were described previously (31).

FIG. 2.

Single amino acid changes in 2B and 3A sequences restore wild-type plaque phenotype to PV-2Cah-HRV14 RNAs. (A) Schematic diagram of chimeric viral transcripts encoding compensatory amino acid changes in either the 2B or the 3A protein. μ2B refers to 2B-T35A and μ3A to 3A-K9R. The sequence encoding the HRV14 N-terminal amphipathic helix (ah) in protein 2C is indicated by a black box, and the engineered amino acid changes in 2B or 3A are indicated by arrows. 5′NTR, 5′ nontranslated region. (B) Plaque morphology of wild-type PV or PV-2Cah-HRV14 with single amino acid substitutions. HeLa cell monolayers were transfected with serial dilutions of the indicated mutant RNA transcripts. Plaques were visualized 48 h after transfection.

FIG. 3.

Accumulation of chimeric PV RNAs in transfected cells. (A) HeLa cell monolayers were transfected with 1 μg of the indicated RNA transcripts. Cells were harvested at the indicated times after transfection. Total cytoplasmic RNA isolated from approximately 5 × 10⁴ cells was used to hybridize with a ³²P-labeled riboprobe complementary to nt 220 to 460 of PV RNA and subjected to digestion with RNases A and T1 as described previously (31). Protected fragments were analyzed by electrophoresis on a 6% polyacrylamide-urea gel. (B) Quantitative presentation of data obtained by analysis of the bands shown in panel A with a Typhoon 8600 imager (Amersham Biosciences) and ImageQuant software.

FIG. 4.

Expression of the GAL4 DNA-binding domain fusion proteins. COS-1 cells were transfected with pBind constructs encoding the indicated GAL4 fusion proteins. At 26 h posttransfection, proteins from total cell lysates were prepared, separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and transferred to nitrocellulose membrane. The blot was probed with monoclonal antibodies against the GAL4 DNA-binding domain, obtained from BD Biosciences.

FIG. 5.

Mammalian cell two-hybrid analysis of interactions of PV fusion proteins. COS cells were cotransfected with pBind-2C (A and B) or pBind-2BC (C and D), and different pAct plasmids that drive expression of 2B, 2C, 2BC, 3A, and 3AB fusion proteins and the reporter vector pG5luc. Luciferase activity was assayed as described in Materials and Methods and is expressed as relative light units (RLU). Values represent means of triplicate transfections, and error bars represent standard deviations. Where no error bars are apparent, the deviations were too small to be visible on the graph. Luciferase expression resulting from transfection of each of the binding partner plasmids alone is shown as a negative control for each protein pair.

FIG. 6.

Effect of the HRV14 amphipathic helix and compensatory mutations in PV protein 2C on protein interactions. (A) Mammalian two-hybrid analysis of the effect of substitutions in N-terminal amphipathic helix in protein 2C on homomultimerization. The cotransfected plasmids expressed wild-type 2C, chimeric 2C with the amphipathic helix from HRV14 (2Cah-HRV), or 2C with the amphipathic helix from HRV with the single amino acid substitution L28F. The reporter vector pG5luc was also cotransfected in each group. Error bars represent standard deviations from triplicate transfections. (B) Mammalian two-hybrid analysis of interactions of mutant forms of 2BC protein with protein 3A. pAct-2BC mutants were cotransfected with pBind-3A and the reporter vector pG5luc into COS cells. RLU, relative light units.

FIG. 7.

Formation of vesicles visualized by electron microscopy in HeLa cells infected with chimeric virus R6 and harvested at 5.5 h (A), infected with R9 and harvested at 6 h (B), or infected with R1 and harvested at 4 h (C). Arrowheads, vesicle clusters; N, nucleus. Bars, 1 μm.